develop

model and weight are not compatible

5

I download the 'dinov2_vitl14_pretrain.pth' locally and use the 'vit_large' function to build model. I run the fellowing code. But the weight and the model are not compatible ```python from dinov2.models.vision_transformer...

Hi, I need the raw images in my research and i submit the google drieve. After that, i receieved the reply of the NLVR^2 datasets, But I don't know how...

Hi, thanks you for your such a great work. I wonder how to split novel and base class when we test on three datasets. I didn't find the implementation. I...

> Sorry, it’s my fault. Thanks a lot. hi, I am also confused about how to download The CIRR dataset, although I have read the README. Can you explain how...

关于相关论文的请教

3

你好，请问又有使用mobilenet替代yolov系列主干和用深度可分离卷积代替普通卷积的相关papper?最近在写本科毕业论文，还要汇报，所以请给些要看论文方面的建议，谢谢

运行get_map时一些问题

1

大佬你好，我使用你提供的主干网络然后训练了自己的权值文件，运行get_map.py 提示会有4952张图片，这个正常吗？ 我有修改了(训练集+验证集)与测试集的比例，然后再次运行voc_annovation.py.产生新的txt,再次运行get_map.py发现还是4952张图片，请问这是怎么回事？

Hi, thanks for your great work. I have a question about the classifier in this paper, I want to know whether you used the 'background' class in the $C_{train}$. >We...

question about 'SD Features'

6

I'm looking forward to seeing you open up the implementation code for the 'SD Features' model mentioned in the paper. Because I think SD features is a research area.

How to retrieval using ComoDiff?

5

Thanks you for porpose this excellent work！ But when I read the paper, I was very confused. When we need to complete the retrieval task, I don't know how to...

How to modify the size of the vocabulary during training ?

1

Hello, thank you for coming up with such a great job. For the content in section 3.2 of this paper: > Training with Online Vocabulary ... The vocabulary for each...